The pitch is straightforward: diesel is expensive, dirty and noisy, so replace it with a battery. Whether that works depends almost entirely on how many hours the generator actually runs and why, and those facts are frequently unknown at the site.
Before any comparison, find out:
Frequent short outages. The best case for storage by a wide margin. A site experiencing thirty brief interruptions a year is starting a large diesel engine thirty times to cover events lasting minutes. Each start causes disproportionate engine wear, and the fuel burned during warm-up is out of all proportion to the energy delivered. A modest battery covers these seamlessly and the generator stops starting altogether.
Rare long outages. The worst case. If the generator runs twice a year for eight hours, sizing a battery to cover that is enormously expensive for an asset that sits idle. The generator is genuinely the right tool – cheap standby capacity, expensive energy, which suits infrequent long events precisely.
Regular peak shaving. Where the generator runs to keep site demand below a connection limit or to avoid demand charges, storage usually wins clearly. The battery does the same job without fuel, without emissions, and it can also arbitrage the tariff when it is not needed for peaks.
Most sites end up keeping the generator and adding storage, which is often the right outcome rather than a compromise.
The battery handles the frequent short events – which is the large majority of them – and covers the first period of a long outage. The generator starts only when the outage extends beyond battery capacity. Generator running hours fall dramatically, servicing intervals stretch, and the site retains genuinely indefinite backup capability.
Storage sized for eight hours of full site load to eliminate a generator is usually a worse investment than storage sized for forty minutes that eliminates 90% of its starts.
On the diesel side: fuel degradation and polishing, statutory testing, fuel storage compliance, emissions permitting, and increasingly, restrictions on running diesel plant in urban areas. Several European cities have tightened this considerably, and future-proofing has value that is hard to quantify but real.
On the battery side: it degrades whether used or not, so a battery held purely for backup is consuming calendar life while earning nothing. This is exactly why stacked applications matter – a battery that shaves peaks daily and provides backup occasionally earns its keep. One that only waits does not.
A packaging plant with a 400 kVA generator running about 55 hours a year across 34 separate events, mostly short. All-in generator cost around 92 per hour, so roughly 5,100 a year, plus a service contract.
Event analysis showed 31 of 34 events lasted under 25 minutes. A 250 kW / 200 kWh battery covers those completely. The generator remains for the three long events.
Direct fuel and maintenance saving was modest – about 4,200 a year. What made the project work was stacking: the same battery shaved demand peaks worth roughly 19,000 annually. The diesel displacement was a secondary benefit that improved the case rather than carrying it.
That is the usual shape. Diesel replacement alone rarely justifies storage. Diesel replacement plus demand management frequently does.